Seminario de Probabilidad y Estadística
Título: Constrained Learning for Dynamical Systems
Expositor: Santiago Paternain (Rensselaer Polytechnic Institute)
Resumen: Learning has shown great success in recent years in controlling complex dynamical systems. However, for the most part, when training a policy most of the algorithms only consider a single objective function. However physical systems are required to satisfy a set of operation constraints, such as safety constraints or minimum performance levels. Naturally, one can express these problems as constrained optimization problems. These problems are in general non-convex and thus challenging. In this talk, I will establish that solving Reinforcement Learning problems under constraints is in fact not harder than solving unconstrained Reinforcement Learning problems.
Viernes 19/11 a las 10:30
zoom
Contacto: Alejandro Cholaquidis - acholaquidis [at] hotmail.com (acholaquidis[at]hotmail[dot]com)
El viernes 3 de Diciembre, realizaremos un almuerzo de cierre del seminario de Probabilidad y Estadística 2021, a las 12.30 en Enriqueta Bar - (Parque Rodó).
Aquellos interesados en asistir, se agradece llenar el siguiente formulario:
***********************************************************************************
Canal de youtube:
https://www.youtube.com/channel/UCOPZEOrLSAYPz2qCAL-KqMg/about
Link de zoom para la charla
https://salavirtual-udelar.zoom.us/j/89466045708?pwd=SENUTDQ3KzZNTWN2U3JScUZIdDZTQT09
ID de reunión: 894 6604 5708
Código de acceso: probable-1